| T42 | <smatkovi> @elros34 libdl_android.so is in these locationshttps://pastebin.com/fvA7FGhS but is expected here somehow: ./out/target/product/j5xnlte/system/apex/com.android.runtime.release/lib/bionic/ | 07:50 |
|---|---|---|
| T42 | <smatkovi> @elros34 libdl_android.so is in these locations https://pastebin.com/fvA7FGhS but is expected here somehow: ./out/target/product/j5xnlte/system/apex/com.android.runtime.release/lib/bionic/ | 07:50 |
| T42 | <elros34> What is latest android version from samsung on your device? What about other libs: libdl.so, libc.so and libm.so? Are they also only in system/lib? | 08:44 |
| T42 | <smatkovi> no they are also in subfolders of SHARED_LIBRARIES in out for example out/target/product/j5xnlte/obj/SHARED_LIBRARIES/libc_intermediates/libc.so | 09:16 |
| T42 | <smatkovi> and there is also lineage os 18 available for it, but camera doesn't work there. or do you mean official release from the manufacturer? | 09:17 |
| T42 | <elros34> yeah from samsung | 09:23 |
| T42 | <elros34> show same search from out/ for those mentioned libs | 09:24 |
| T42 | <smatkovi> 7.1.1 | 09:27 |
| T42 | <smatkovi> https://pastebin.com/J3L1fMHv | 09:29 |
| T42 | <smatkovi> maybe it is because i didn't build the kernel with selinux | 09:32 |
| T42 | <smatkovi> i'm trying now but it says unknown variable '$(PATH_OVERRIDE_SOONG)' | 09:32 |
| T42 | <smatkovi> i had that already and don't know if only make clean helped here or something else | 09:33 |
| T42 | <elros34> no | 09:33 |
| T42 | <elros34> also I told you to search only in out/ not in whole system... | 09:33 |
| T42 | <smatkovi> oh sry | 09:33 |
| T42 | <elros34> remove "dl_android" from here: https://github.com/mer-hybris/droid-hal-device/blob/master/droid-hal-device.inc#L658-660 | 09:37 |
| T42 | <smatkovi> thanks i'm trying now | 09:40 |
| T42 | <smatkovi> shouldn't i rebuild the kernel with selinux enabled? | 09:40 |
| T42 | <elros34> wasn't it enable by default? | 09:41 |
| T42 | <smatkovi> at least CONFIG_SECURITY_SELINUX=y wasn't in the defconfig before, i don't know if it could be enabled somewhere else | 09:42 |
| T42 | <elros34> was it in .config? | 09:43 |
| T42 | <smatkovi> sorry which config? | 09:44 |
| T42 | <elros34> .config file in out/. That file which you should check via mer-kernel-check according to hadk pdf | 09:45 |
| T42 | <smatkovi> i see, would it show up as error or warning here? | 09:50 |
| T42 | <smatkovi> okay warning WARNING: CONFIG_SECURITY_SELINUX_BOOTPARAM_VALUE is invalid | 09:51 |
| T42 | <smatkovi> WARNING: CONFIG_SECURITY_SELINUX_BOOTPARAM is invalid | 09:51 |
| T42 | <smatkovi> i guess CONFIG_SECURITY_SELINUX_BOOTPARAM should be set to y. should also CONFIG_SECURITY_SELINUX be set to y? | 09:55 |
| mal | which device is that again? | 09:56 |
| T42 | <smatkovi> j5xnlte or samsung galaxy j5 2016 | 09:57 |
| T42 | <smatkovi> oh breakfast brings in this environment variable. rebuilding the kernel now and reporting back afterwars | 10:08 |
| mal | you always need to run those source and breakfast commands before building when you re-enter the build environmen | 10:14 |
| mal | I mean before running make hybris-hal for example for android side build | 10:14 |
| T42 | <elros34> SECURITY_SELINUX_BOOTPARAM* is irrelevant for you. About CONFIG_SECURITY_SELINUX: again check .config file, I bet it was already enabled | 10:16 |
| T42 | <elros34> basically if you do not touch anything selinux related then it's probably all correct | 10:17 |
| T42 | <smatkovi> okay thanks now it is enabled correctly | 11:14 |
| T42 | <smatkovi> okay now it gives the same error as when i copied the file there File not found: /home/sebastian/hadk/installroot/default.prop i found a default.prop and put it there but it gets overwritten everytime i build droid-hal | 11:22 |
| T42 | <smatkovi> or more correctly the directory gets deleted before building | 11:22 |
| T42 | <elros34> if you do not have default.prop somewhere in out/ then remove it from droid-hal-device.inc | 11:25 |
| T42 | <elros34> you can't put files in installroot/ it's cleared/created during rpm creation | 11:26 |
| T42 | <smatkovi> YES: * Building of droid-hal-j5xnlte finished successfully | 11:35 |
| T42 | <smatkovi> nice thanks | 11:35 |
| T42 | <smatkovi> okay now i have some missing dependencies in building —config | 11:53 |
| T42 | <smatkovi> community-adaptation is needed by droid-config-jmxnlte-1-202307011146.armv7hl | 11:53 |
| T42 | <smatkovi> pkgconfig(android-headers) is needed by droid-config-jmxnlte-1-202307011146.armv7hl | 11:53 |
| T42 | <smatkovi> ssu-kickstart-configuration is needed by droid-config-jmxnlte-1-202307011146.armv7hl | 11:53 |
| T42 | <smatkovi> are they installed or built before that? | 11:53 |
| T42 | <smatkovi> but this is also strange: Building target platforms: armv7hl-meego-linux | 11:54 |
| T42 | <smatkovi> Building for target armv7hl-meego-linux | 11:54 |
| T42 | <smatkovi> shouldn't it be a target with the device name? | 11:54 |
| T42 | <elros34> run script without any arguments | 11:57 |
| T42 | <smatkovi> okay so i'm running this now rpm/dhd/helpers/build_packages.sh | 11:57 |
| T42 | <smatkovi> unfortunately the same. maybe it's because of these warnings before the error? https://pastebin.com/5gr9svUj | 12:00 |
| T42 | <elros34> read whole log, not just few lines which in fact are irrelevant | 12:02 |
| T42 | <smatkovi> https://pastebin.com/mKZUguJL | 12:05 |
| T42 | <elros34> first error in log: " DB_VERSION_MISMATCH: Database environment version mismatch". Solution is available in channel logs and even in hadk-hot | 12:06 |
| mal | "sb2 -t $VENDOR-$DEVICE-$PORT_ARCH -R -m sdk-install rpm --rebuilddb" | 12:08 |
| T42 | <smatkovi> it seebs you provided the solution: sb2 -t <target> -m sdk-install -R rm -rf /var/lib/rpm/__db.00* | 12:10 |
| T42 | <smatkovi> okay thanks mal i will try your command first | 12:10 |
| mal | both probably work, the one I just gave is cleaner way | 12:10 |
| T42 | <smatkovi> okay thanks. now i'm unsure the package is successfully built but there are some errors about bluez missing maybe i can proceed ignoring those?: https://pastebin.com/Y3L8fGpn | 12:13 |
| T42 | <smatkovi> at building mw it asks to build libhybris, so i said yes to all | 12:15 |
| mal | that error in config package build is strange | 12:17 |
| T42 | <smatkovi> yes version also doesn't build successfully: https://pastebin.com/KHXtDnwt | 12:24 |
| T42 | <elros34> do you see which line is the problemthe problem | 12:26 |
| T42 | <smatkovi> maybe the ofono-configs conflict? | 12:28 |
| T42 | <smatkovi> i could delete ofono-configs-binder | 12:28 |
| T42 | <smatkovi> and keep droid config-jmxnlte | 12:29 |
| T42 | <elros34> great find but now search in existing docs you have for correct solution | 12:29 |
| T42 | <elros34> ctrl+f and you will find it faster then you can write about it in this chat | 12:30 |
| mal | do you have these 4 lines https://github.com/mlehtima/droid-config-fp4/blob/devel/rpm/droid-config-fp4.spec#L20 | 12:31 |
| mal | in droid-config spec file | 12:32 |
| T42 | <TheVancedGamer> mal: recalling yesterday, do you have any idea about that media.c2 issue? | 12:32 |
| mal | @TheVancedGamer not right now, maybe I should try to get my latest android 13 based build running again | 12:34 |
| T42 | <smatkovi> no i only added %define android_version_major 10 | 12:34 |
| T42 | <TheVancedGamer> mal: i'll wait then, in the meanwhile I was trying to figure out why it requests c2 but sadly no progress | 12:35 |
| T42 | <smatkovi> but i added these 4 lines now | 12:35 |
| T42 | <smatkovi> running the buildscript without arguments gives me this in the config log https://pastebin.com/LMnnJuhA | 12:46 |
| T42 | <elros34> typos in spec file? | 12:48 |
| T42 | <smatkovi> i had one where i wrote j5xnlet instead of j5xnlte but that was earlier, yes maybe there are more... | 12:49 |
| T42 | <elros34> I am talking about your latest changes, do you read logs you put here? "error: File must begin with "/": ofono-configs" | 12:50 |
| T42 | <smatkovi> yes but i think that is strange | 12:53 |
| T42 | <smatkovi> it was without / in the linked spec file | 12:54 |
| T42 | <smatkovi> should i just put / infront of every ofono-configs*? | 12:54 |
| T42 | <elros34> error may be misleading but it indicates some mistake in file related to that change | 12:54 |
| T42 | <smatkovi> okay i changed these lines to: # Device-specific ofono configuration | 12:56 |
| T42 | <smatkovi> Provides: droid-configs-j5xnlte | 12:56 |
| T42 | <smatkovi> Obsoletes: ofono-configs-binder | 12:56 |
| T42 | <elros34> no, show your changes if you do not know what is wrong | 12:57 |
| T42 | <smatkovi> i was asked if i have these 4 (3) lines: Provides: ofono-configs | 12:58 |
| T42 | <smatkovi> Obsoletes: ofono-configs-mer | 12:58 |
| T42 | <smatkovi> Obsoletes: ofono-configs-binder | 12:58 |
| T42 | <smatkovi> and i just added those | 12:58 |
| T42 | <elros34> Then show the file | 12:59 |
| T42 | <smatkovi> so this section# Device-specific ofono configuration | 12:59 |
| T42 | <smatkovi> ok here is my spec file https://pastebin.com/eyUjcqmy | 13:00 |
| T42 | <elros34> remove everything after: %include patterns/patterns-sailfish-device-configuration-j5xnlte.inc and then add these 3 lines before %include droid-configs-device/droid-configs.inc | 13:02 |
| T42 | <smatkovi> okay thanks | 13:04 |
| T42 | <smatkovi> cool it seems to have built it. i wonder if setting RELEASE variable in .hadk.env is mandatory | 13:12 |
| T42 | <smatkovi> okay it seem to work to just export it | 13:15 |
| T42 | <smatkovi> i get these errors building mic grep: /home/sebastian/hadk/hybris/droid-configs/installroot/usr/share/kickstarts/Jolla-@RELEASE@-j5xnlte-@ARCH@.ks: No such file or directory | 13:15 |
| T42 | <smatkovi> sed: can't read /home/sebastian/hadk/hybris/droid-configs/installroot/usr/share/kickstarts/Jolla-@RELEASE@-j5xnlte-@ARCH@.ks: No such file or directory | 13:15 |
| T42 | <smatkovi> do i have to write a kickstart file? | 13:15 |
| T42 | <smatkovi> i don't know how to do that | 13:15 |
| T42 | <elros34> I see some mismatch jmxnlte and j5xnlte, make sure you use correct name everywhere | 13:20 |
| T42 | <smatkovi> where do you see that? | 13:22 |
| T42 | <elros34> all over your logs | 13:22 |
| T42 | <elros34> even in your latest paste: %define device jmxnlte and device-adaptation-j5xnlte.inc. WHich on eis correct | 13:22 |
| T42 | <smatkovi> oh thanks | 13:23 |
| T42 | <smatkovi> j5xnlte | 13:23 |
| T42 | <smatkovi> i have a zip file :) | 13:41 |
| T42 | <smatkovi> flashing with twrp it says failed to extract filesystem. maybe the zip file isn't correctly copied onto the disk from that i install it? | 13:49 |
| T42 | <smatkovi> i will take a break now and see if i can flash it later | 13:49 |
| T42 | <smatkovi> and i also get error 7 after failed to extract filesystem | 14:14 |
| T42 | <smatkovi> maybe my fixup-mountpoints is wrong? | 14:15 |
| T42 | <neochapay> ``` | 14:30 |
| T42 | <neochapay> Jul 01 02:58:21 T1100 sh[982]: nohup: can't execute '/sbin/droid-hal-init': No such file or directory | 14:30 |
| T42 | <neochapay> ``` | 14:30 |
| T42 | <neochapay> i think it's not good... | 14:30 |
| T42 | <elros34> probably missing /system | 14:31 |
| Mister_Magister | any clues on debugging suspend when in dmesg theres not even an indication of trying to go into suspend? | 15:54 |
| Mister_Magister | rinigus: ping | 15:56 |
| rinigus | pong. re question above - check that there is no wakelock up. | 15:57 |
| rinigus | if it is still valid these days | 15:57 |
| T42 | <Mister_Magister> how do i do that? I just pinged you cause you seemed to ask "looks like CONFIG_PM_AUTOSLEEP is deprecated" in android 10 which I have android 11 | 15:58 |
| T42 | <TheVancedGamer> isn't it something like /sys/power/wake_lock | 15:59 |
| T42 | <TheVancedGamer> and /sys/power/wake_unlock | 15:59 |
| T42 | <Mister_Magister> sry i never dealed with sleep issues before | 15:59 |
| T42 | <Mister_Magister> /sys/power/wake_lock seems to return dsme_iphb_wakeup | 16:00 |
| T42 | <Mister_Magister> whatever that is | 16:00 |
| T42 | <Mister_Magister> unlock has dsme_iphb_wakeup dsme_rtc_input hal_bluetooth_lock lowirpcd:6284 mce_mux mce_rtc_wakeup psensor_domain rmt_storage_507833830592 rmt_storage_507834870976 rpmb_access_wakelock sensorfwd_pass_data sscrpcd:5552 tftp_server_wakelock usb_moded_input usb_moded_state | 16:00 |
| T42 | <TheVancedGamer> try doing `echo dsme_iphb_wakeup > /sys/power/wake_unlock` for funsies | 16:00 |
| T42 | <TheVancedGamer> and see if it suspends | 16:00 |
| T42 | <Mister_Magister> i mean its already in unlock but okai | 16:01 |
| T42 | <Mister_Magister> yep it didn't scream at me | 16:01 |
| T42 | <Mister_Magister> now its only in unlock | 16:01 |
| T42 | <TheVancedGamer> try and see if it sleeps lol | 16:01 |
| T42 | <TheVancedGamer> if not I forgot how this works | 16:01 |
| T42 | <Mister_Magister> nope, doesn't seem to be sleeping | 16:03 |
| T42 | <Mister_Magister> but ill try without wifi | 16:03 |
| T42 | <Mister_Magister> nope no suspend without wifi either | 16:04 |
| T42 | <Mister_Magister> biggest problem is that there's absolutely no info in dmesg about suspend, not even failign | 16:15 |
| T42 | <Mister_Magister> any clues welcome | 16:18 |
| T42 | <Mister_Magister> i'll try to enable autosuspend | 16:19 |
| T42 | <elros34> does it sleep at all? Systemdatascope shows nice graphs including sleeping | 16:20 |
| T42 | <Mister_Magister> not at all | 16:20 |
| T42 | <Mister_Magister> i'm using systemmonitor | 16:20 |
| T42 | <Mister_Magister> it has -0% sleep and like nothing in dmesg at all not even trying not even failing | 16:21 |
| Mister_Magister | it doesn't even try to fail | 16:22 |
| T42 | <elros34> maybe kill this related process hal_bluetooth_lock? You didn't disable sleep in mce and forget about it:P | 16:30 |
| T42 | <Mister_Magister> nah i don't think sleep was disabled in mce | 16:31 |
| T42 | <Mister_Magister> what did hal_bluetooth_lock do? lol | 16:32 |
| T42 | <elros34> who knows but I think wifi/bluetooth are first to eliminate | 16:34 |
| T42 | <Mister_Magister> i mean they're off | 16:34 |
| T42 | <Mister_Magister> bt is off | 16:34 |
| T42 | <Mister_Magister> i'll try enabling the pm_autosuspend | 16:38 |
| rinigus | @Mister_Magister: do you have autosuspend enabled in kernel? I think in android they may suspend somewhat differently now. but you have to search on what is used now if kernel autosuspend will refuse to work | 16:42 |
| rinigus | I think it was something in userspace that was suspending | 16:42 |
| T42 | <Mister_Magister> ye like i said i'm gonna try enabling it so its disabled :P | 16:42 |
| rinigus | as android doesn't use kernel autosuspend anymore, you may get some weird bugs as well. that was a problem that I had with sony tama. | 16:43 |
| T42 | <Mister_Magister> thats why i pinged u cause u said its deprecated but its still there lol | 16:43 |
| T42 | <Mister_Magister> ah | 16:43 |
| T42 | <Mister_Magister> but i should be using it right? | 16:43 |
| rinigus | as far as I know, yes, you should in SFOS | 16:43 |
| rinigus | let's hope it works for you | 16:44 |
| T42 | <Mister_Magister> aight | 16:45 |
| T42 | <Mister_Magister> it takes solid 30 minutes to rebuild this trashy kernel so gimme a moment | 16:45 |
| T42 | <Mister_Magister> now droid-hal doesn't want to build letsgo | 16:56 |
| T42 | <Mister_Magister> aaand no suspend with autosuspend either | 17:14 |
| rinigus | @Mister_Magister: pity. | 17:18 |
| T42 | <Mister_Magister> ill try suspend patch i had on 5z in android 9 | 17:18 |
| T42 | <Mister_Magister> first need to figure out why droid-hal doesn't build properly | 17:20 |
| T42 | <Mister_Magister> ayy wiat withuot wifi there is sleep | 18:14 |
| T42 | <Mister_Magister> 83% so not great but there is something | 18:14 |
| T42 | <Mister_Magister> ayy after patch got some wifi suspend too | 18:25 |
| T42 | <Mister_Magister> /sys/power/wake_lock has only mce_mux mce_pwrkey_stm sensorfwd_pass_data | 18:25 |
| T42 | <Mister_Magister> tho with wifi there's lot of suspend failed | 18:26 |
| T42 | <Mister_Magister> but its a start | 18:26 |
| T42 | <Mister_Magister> btw any clues how to stop wpa_supplicant from spamming dmesg? https://paste.opensuse.org/pastes/2ba394484b5e | 18:36 |
| T42 | <smatkovi> bard says: | 19:08 |
| T42 | <smatkovi> /etc/wpa_supplicant/wpa_supplicant.conf file: | 19:08 |
| T42 | <smatkovi> Code snippet | 19:08 |
| T42 | <smatkovi> ctrl_interface=DIR=/var/run/wpa_supplicant GROUP=netdev | 19:08 |
| T42 | <smatkovi> or You can also add the following line to your /etc/default/wpa_supplicant file: | 19:08 |
| T42 | <smatkovi> quiet=1 | 19:08 |
| T42 | <Mister_Magister> ah thank | 19:10 |
| T42 | <smatkovi> but beware that this could be wrong too | 19:11 |
| T42 | <Mister_Magister> huh? | 19:11 |
| T42 | <smatkovi> it's a response from a large language model from google | 19:12 |
| T42 | <Mister_Magister> ew | 19:12 |
| T42 | <Mister_Magister> thats 99.99999% wrong | 19:12 |
| T42 | <Mister_Magister> scratch that make it 100 | 19:13 |
| T42 | <Mister_Magister> in 95% of cases any language model is just plain wrong | 19:14 |
| T42 | <smatkovi> you can try it and then tell us if one of these "solutions" worked | 19:18 |
| T42 | <smatkovi> so if i get failed to extract filesystem and error 7 when flashing sailfish os it could be that system is missing? | 19:29 |
| T42 | <Mister_Magister> no | 19:30 |
| T42 | <Mister_Magister> error 7 means you mostlikely don't have bzip2 | 19:30 |
| T42 | <Mister_Magister> hence it can't extract | 19:30 |
| T42 | <smatkovi> so twrp might not have bzip2? | 19:49 |
| T42 | <Mister_Magister> no | 19:49 |
| T42 | <Mister_Magister> it DOESNT have it | 19:49 |
| T42 | <Mister_Magister> thats not even a question, newer twrp just doesn't have it | 19:49 |
| T42 | <Mister_Magister> So about the suspend. Without wifi i'm getting steady 83% sleep without fails, with wifi, i'm getting also around 83% but with some fails | 20:48 |
| T42 | <Mister_Magister> should i worry? iirc it should be more like 95-99% sleep | 20:48 |
| T42 | <Mister_Magister> ye my g2 gets close to 100% sleep | 20:48 |
| T42 | <smatkovi> thanks @mister_magister i extacted it to a tar and changed the scripts accordingly, now i can flash it | 21:56 |
| T42 | <Mister_Magister> np | 21:56 |
| T42 | <smatkovi> unfortunately i get kernel is not seandroid enforcing before even booting, so it must have to do with selinux i guess | 21:57 |
| T42 | <smatkovi> i will try by setting selinux to permissive | 22:09 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!